polarization opなし → matrix(引数だけで決まる op —— 画像やデータの入力を取らない)import fullseye as fs; fs.ledger.mueller_partial_polarizer(angle_deg=0.0, extinction_ratio=1000.0, transmission=1.0) (実装を直接呼ぶなら import optics; optics.mueller_partial_polarizer(angle_deg=0.0, extinction_ratio=1000.0, transmission=1.0)、台帳から引くなら opsoptics.get("mueller_partial_polarizer"))Mueller matrix of a linear polariser with a finite extinction ratio (a linear diattenuator). → (4, 4), or (*shape, 4, 4) for an angle field.
A real polariser passes intensity transmittance T1 = transmission for
light polarised along its axis and T2 = T1 / extinction_ratio for light
across it. Writing P = T1 + T2, D = T1 - T2, Q = 2 sqrt(T1 T2)
and c = cos 2a, s = sin 2a (the axis at angle_deg), the matrix is
``0.5 * [[P, D c, D s, 0],``
`` [D c, P c^2 + Q s^2, (P - Q) c s, 0],``
`` [D s, (P - Q) c s, P s^2 + Q c^2, 0],``
`` [0, 0, 0, Q]]``
— the textbook diattenuator diag-block form rotated to angle_deg.
It is also exactly the Mueller matrix of the Jones matrix
R(a) diag(sqrt T1, sqrt T2) R(-a) (a pure diattenuator has no
retardance and no depolarisation), and the tests check that equality
through the generic Jones → Mueller conversion, a separate code path.
Ground truth (closed form, checked in the tests): unpolarised light comes
out with intensity (T1 + T2)/2 and degree of polarisation
(ER - 1)/(ER + 1); the intensity transmitted through two such
polarisers, parallel vs crossed, is (T1^2 + T2^2)/2 vs T1 T2 —
the measured extinction ratio of the pair is (ER^2 + 1)/(2 ER), not
ER; Malus’s law becomes T2 + (T1 - T2) cos^2(theta) for fully
polarised input at angle theta to the axis; ER -> infinity tends to
T1 * :func:mueller_element ("polarizer"); ER = 1 is a
neutral-density filter T1 * I.
angle_deg may be an array (one matrix per entry, as
:func:mueller_element); extinction_ratio and transmission are
scalars. Angles follow the module convention (from +x toward +y = down the
rows, clockwise on a displayed image).
Raises ValueError: non-finite / non-real / bool / string arguments;
extinction_ratio < 1 (an “extinction ratio” below 1 means the axis is the
other one — rotate by 90 degrees instead of passing the reciprocal) or
infinite (use :func:mueller_element for the ideal polariser);
transmission outside (0, 1] (a passive element cannot amplify, and a
zero-transmission element has no axis); an angle field over
:data:MAX_FIELD_ELEMENTS / 16 entries.
optics の全 op は入力を検証してから計算する(黙って通さない):
_mm / _um / _deg / _mrad。mm と µm の取り違えは crash ではなく「もっともらしく間違った答え」なので、名前で防ぐ。大きさから単位を推測する処理は一切しない。ValueError — float('50') は成功してしまうため、未パースの設定値が長さとして通り抜ける(実測: thin_lens('50', '200') がもっともらしい 66.667 mm を返していた)。bool も True == 1 の暗黙昇格として拒否。ValueError(実数枠のみ。虚部の無言切り捨て・マスク剥がしを拒否)。NaN/Inf は全入力で ValueError。depth_of_field の過焦点距離以遠の far_mm = inf(それが過焦点距離の定義)と gaussian_beam のウエストでの wavefront_radius_mm = inf(平面波面の曲率半径)。どちらも有限の相棒(far_is_infinite / curvature_per_mm)を併せて返す。それ以外の無言 NaN/Inf は内部で検出して ValueError —「float64 が溢れた」と「答えが無限大」は別の主張なので、後者の顔で前者を返さない。optics.MAX_GRID(4096)、供給された場/PSF/開口は optics.MAX_FIELD_ELEMENTS(2^24)、ABCD 素子列は optics.MAX_SYSTEM_ELEMENTS(1024)、Zernike は MAX_ZERNIKE_TERMS(512)/ MAX_ZERNIKE_ORDER(40)/ MAX_ZERNIKE_BASIS(2^25)。小さな引数から巨大な内部確保が起きる経路(実測: n_max=40 × 4096² で 108 GB)を fail-closed で塞ぐ。| 物理的に不可能な状態も拒否: 偏光度 > 1 の Stokes ベクトル、負の透過率、負の強度、n- | m | が奇数などの不正な Zernike 添字。 |
py -3.11 examples/polarization_camera_pipeline.pymatrix を入力に取れる)abcd_trace · emva_filter_distortion · mueller_apply · mueller_checks · equirect_rotate · pixel_shift_capture · pixel_shift_merge · pixel_shift_motion_mask
polarization)jones_element · jones_apply · stokes_from_jones · mueller_element · mueller_apply · stokes_analyze · polarization_demosaic · polarization_demosaic_color
Provenance: optics.py — OPTICS operator registry. この per-op ノートは tools/opdocs.py md が自動生成(手編集しない)。
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.